The detection of biologically important sequence variation is a major goal of modern cancer genomics. In an effort to focus resequencing investigations on high-value regions, many researchers have chosen to focus on the exome (protein-coding exons), which is expected to be enriched for functional variation. In the near future, comprehensive resequencing of an individual's exome may cost less than $1000, substantially less than whole-genome sequencing. We anticipate that these developments will yield an enormous amount of exonic sequence variation in cancer genomes that will provide leads to new cancer genes and drug targets. Most of this variation is likely to be effectively neutral, thus the cancer genomics community will have to identify and prioritize those variants that warrant further studies in assay systems, e.g. cell lines and model organisms. At this time, there are no publicly available resources that enable researchers who are not bioinformatics experts to explore the biological importance of sequence variants discovered in their investigations. There is a pressing need for tools that enable transparent, researcher-driven exploration of variation, by mapping variants onto proteins and pathways and presenting results in a way that is accessible to the average researcher. Development of such resources will allow rapid translation of genomic investigations into tangible progress in medical research. We propose here to develop a novel computational application tailored to the needs of researchers who are discovering variation in the exomes of cancers. Our work will produce a high- throughput annotation pipeline, with associated web-based analysis and visualization tools. This resource will enable users to interpret and prioritize tumor-derived sequence changes and help them think about how to design functional tests for variants of interest. We will demonstrate the utility of the application by collaborating with researchers from the Johns Hopkins Sidney Kimmel Cancer Research Center, who are sequencing the exomes of eleven cancer types during the two years for which we request funding (approximately 23 primary tumors per cancer type). The application will be used in collaboration with this team of researchers, led by Drs. Bert Vogelstein, Ken Kinzler, and Victor Velculescu to identify genes responsible for susceptibility to and progression of these cancers (Support Letter attached). This work will lay the foundation for a broader application of the tools to exomic variation data in additional cancers, such as those being studied by NCI's Cancer Genome Atlas project and TARGET.